Kenilworth Home Management Platform

Kenilworth Home Management Platform

Published in 2021

Shortly after we had our first child, I became acutely aware of the logistical challenges of modern parenting. And as a tool maker, I set out to create a solution that smoothed out some of the pain points.

I approached the project as a platform that could eventually span multiple surfaces – so a robust API layer would be key. I wanted it to reduce my wife and my mental loads, and serve as a central repository for key knowledge related to our family and home.

The result was the Kenilworth app. Features included:

  1. Meal Planner – The keystone feature. This allowed us to plan ahead for the week, embracing the idea of “recognition over recall” and allowing selection from all of our local restaurants and favorite recipes. It allowed for levels of certainty and optionality with Placeholders: Some days, you just want to decide last minute, and that’s ok.
  2. Restaurants – A simple repository for restaurants, serving the Meal Planner feature, which delivery service they’re available on, whether or not it’s a chain (we typically prefer local shops), and what cuisine it is.
  3. Recipes – Another repository, serving the Meal Planner, that linked out to favorite recipes. If it was a verbal recipe, we could log it down as a note without a link.
  4. Manuals – I’m a huge fan of RTFM: “Read The F’ing Manual”. And no less for all household appliances. This was a place to centralize all of our digital appliance manuals, in addition to the model names and numbers for service repairs.
  5. Swatches – A lowkey pain point for all households. This was a place to log the paint codes, types, and brands for the paint used for each room in our house. Incredible how difficult this is to codify and manage.
  6. Home Projects – A place to manage and triage the many home projects that crop up. This included Priority, Expected Cost, Actual Cost, and Service Providers.
  7. Subscriptions – It seems like everything is a paid subscription these days, and for many people it’s caused expenses to get quite a bit out of control. This feature allowed families to track the active and recurring statuses of subscriptions, the login method, cost, and category. Categories could be rolled up to help with budgeting and planning.
  8. Guest Book – A separate, externally-facing experience for when guests visit and want to leave a note. This experience also provided things like the wifi password, local attractions, favorite restaurants, and more.
  9. Dashboard – A lens into trends and unique views of the data, such as “how often are we eating leftovers?” or “How often are we eating from restaurants vs home-cooked recipes?” It also served as a bookmark hub for couples to centralize shared documents and savings/spending goals.

This existed and evolved for about 3 years as an increasingly sophisticated React + Redux web app. I dabbled in different experiences for Slack, Apple TV, Apple Watch, and an iPhone app with widgets, all served by a robust API layer written on Java Spring.

The implementation was entirely WCAG AA compliant, keyboard accessible, and responsive for mobile usage.

Architecture & API

I learned a LOT about authentication, API design, and data state management with Redux. The API leveraged inherited classes, interfaces, and structs for a full-featured OOP Java implementation. Here are a couple of views into the architectural planning:

Marketing Home

As this became more fully formed, I thought my local community might be able to benefit from it at some point. I think it’s super important to be a good neighbor, and this was one potential tool that could help my community and neighbors. I didn’t see this as a commercial project but rather as a dutiful pro-bono service for my community.

With that in mind, here was the signup homepage that almost was, until…

Enter ChatGPT

Things changed once ChatGPT entered the picture around late 2022 – I immediately knew things were different. (Em-dash pun intended. 😉)

I tested the potential by plugging exported Restaurant and Recipe data into an agent, along with additional recommendation criteria, and it produced a perfectly arranged schedule for our family for the week.

Additionally, my wife has a strong preference for spreadsheets and Apple Notes, and nothing could persuade her otherwise.

Current Status

These two factors encouraged me to put a hold on this project for the time being. For now, Kenilworth has dissolved into the fleeting, etherial waters of Google Spreadsheets and agentic prompts.